Peng Sun

Unified Continuous Generative Models for Denoising-based Diffusion

APEX: One-Step High-Resolution Image Synthesis

Safeguarding Visual Privacy in Dataset Distillation: Robust Initialization via Augmentation

TwinFlow: Realizing One-step Generation on Large Models with Self-adversarial Flows

Any-step Generation via N-th Order Recursive Consistent Velocity Field Estimation

Generative Modeling with Explicit Memory

Beyond Model-Centric: Collaborative Data Optimization for Reusing and Sharing

Efficient Generative Model Training via Embedded Representation Warmup

IOMM: Fast Pre-training of Unified Multimodal Models without Text-Image Pairs

Optimizing Decoding Paths in Masked Diffusion Models by Quantifying Uncertainty

Toward Editable Vector Graphics: Layered SVG Synthesis from Multimodal Prompts

Atomos: Hierarchical Reasoning From Atomic Steps

Structured-Initialization Learning

GIFT: Unlocking Full Potential of Labels in Distilled Dataset at Near-zero Cost

Equally Critical: Samples, Targets, and Their Mappings in Datasets

Privacy as a Free Lunch: Crafting Initial Distilled Datasets through the Kaleidoscope

Collaborative Data Optimization

Co-Evolution Learning

Data-Evolution Learning